The present application relates generally to an improved data processing apparatus and method and more specifically to an apparatus and method for security screening image analysis simplification through object pattern identification.
Operators of X-ray scanners are faced with the difficult problem of identifying threats and ignoring mundane items. Carry-on bags are typically littered with many consumer electronics that add to the visual clutter in an X-ray image and may mask true threats or cause false positives. They also cause mental fatigue in the operators who are typically limited to fifteen minutes of viewing shift.
Current scanners may perform X-ray scans in three dimensions. Scanners may also color code materials. For example, a scanner may assign different colors for organic materials, plastics, metals, etc.
The goal of a security organization is to scan thousands of bags an hour with low false positives and 100% identification of threats. Currently, threat objects are partially automatically identified by fuzzy logic pattern matching, where a number of bags are placed through the system. The scanning system is trained to recognize likely patterns of groupings of objects, because threats may be disassembled in the bag or may be spread across a plurality of bags across a period of time.